Abstract: Synthesized-beam RFID readers may be used to manage and provide information about RFID tag populations. In one embodiment, two or more synthesized-beam readers synthesize respective RF beams towards a tag location. The synthesized-beam readers may coordinate their pointing by means of a controller, a peer-to-peer network, or by using a master-slave arrangement. The synthesized-beam readers may coordinate their transmissions to increase the RF energy available to a tag at the pointing location.

Abstract: A card reader to read information in an integrated circuit (IC) card includes a housing, a first antenna, a control circuit, and a switch unit. The first antenna is disposed inside of the housing to communicate with the IC card. The control circuit communicates with the IC card within a communication range of the first antenna and performs at least a process that acquires data from a memory in the IC card. The switch unit is disposed in the housing. The switch unit includes a radio frequency (RF) tag having a second antenna. The switch unit functions as a switch that performs a predetermined function by causing the RF tag to communicate with the control circuit via the second antenna and the first antenna when a predetermined condition is satisfied.

Abstract: A radio frequency identification (RFID) network and methods for even monitoring in a closed environment includes distributing RFID readers across the environment. Readers have unique reader identifiers and communicate with a computer system. These identifiers and reader locations are stored. The location of subjects bearing passive RFID tags is acquired by reading the unique subject identifiers off the tags and associating these RFID tags with the reader identifiers of the readers making the readings. A subject data store includes the location data of the subjects obtained by the RFID readers and electronic addresses of the subjects. An instruction set data store comprising sets, each set corresponding to one of a plurality of events, is also maintained. In accordance with the instruction set data store, in response to an event, different event messages are sent to different subjects and an aerial drone is dispatched.

Abstract: A method for managing mobile assets includes initially certifying a physical existence and location of a mobile asset. The certifying includes receiving certification information about the location and data representing the physical existence of the mobile asset into a database. After certification, interrogation information is received into the database. The interrogation information includes read information from an electronically readable tag associated with the mobile asset, user-entered information about at least one of an instant location of the mobile asset and user-entered data representing the physical existence of the mobile asset, and geolocation information of the reader. At least one of the instant location and the user-entered data representing the physical existence of the mobile asset is validated by cross-referencing the certification information and the interrogation information to identify any discrepancies.

Abstract: An image capturing device includes an image forming device including a sensor defining an optical reception axis, at least one reading distance, and a region framed by the sensor on a substrate at said at least one reading distance. An illumination device includes an array of adjacent light sources defining an optical illumination axis. The light sources are individually drivable, and each light source is adapted to illuminate an area of a size much smaller than the size of said region framed by the sensor. The illumination axis does not coincide with the reception axis. A driver of the light sources is adapted to drive the light sources so as to switch off at least the light sources that illuminate outside of the boundary of the region framed by the sensor on the substrate at said at least one reading distance.

Abstract: A scanning system has a collimated light source and a movable reflective surface. The surface receives light from the light source, reflects the received light to a target surface, receives light reflected from the target surface, and reflects this light to a detector. An aperture disposed between the reflective surface and the target surface has a non-transmissive portion with a width in a direction perpendicular to the laser scan direction that varies over the scan direction.

Abstract: A hand-held indicia-reading device includes a housing and a cable having a unitary-construction. The indicia-reader has a head portion with an indicia-reading system including a printed circuit board. The entire handle portion of the reader is typically over-molded onto a host-connector cable. The host-connector cable includes a circuit board connector on one end and a connector for connecting to a host at an opposite end extending from a base portion of the handle.

Abstract: Techniques are disclosed for light-based communication using a passive light-reflective device having specially coded reflective or printed optics. The optics can be mounted to an object and configured to reflect light such that a receiver is able to receive the reflected light. The optics are further configured to alternatively display a number of different patterns that change as the receiver moves with respect to the optics, thus causing the receiver to receive an apparent stream of modulated light, which represents coded information that can be decoded into meaningful information. The optics can be mounted to a traffic control or other roadside device. As a vehicle approaches and passes the traffic control device, light reflects off of the optics in a series of patterns. This reflected light can be received by the vehicle and processed to relay the information to the operator or on-board vehicle system.

Abstract: In measurement data selection, when a vector composed of measurement data obtained for respective combinations of light emitting positions, light detection positions, and resolving times in a time-resolved waveform is given as y, a vector in which pixel values of learning image data are components is given as x, and a system matrix for calculating internal image data from the measurement data is given as A1, the vector y which meets the conditional expressions (2) and (3) min?y?00??(2), ?x?A1Ty?22??2??(3) or the conditional expression (4) min(?y?00+??x?A1Ty?22)??(4) is determined, and upon measurement of a subject, only the measurement data corresponding to nonzero components of the vector y is used to prepare the internal image data.

Abstract: There is described an apparatus for outputting a group of images for display, the group being taken from a plurality of images, each image in the plurality of images having a face located therein, the apparatus comprising a processor configured to: retrieve a set of images from a storage medium, the set of images containing at least the plurality of images from which the group of images to be displayed is selected; identify the face in each of the plurality of images; identify variable features on the face in each of the plurality of images; establish the group of images in accordance with a measure of the dissimilarity between the variable features in the plurality of images; and output the group of images for display.

Abstract: A method of processing an image involves detecting a face from a first image; determining a composition of the first image; selecting a composition of a second image according to the composition of the first image; and generating the second image including the face by trimming the first image, according to the composition of the second image.

Abstract: The present invention relates to a method and apparatus for standoff facial and ocular acquisition. Embodiments of the invention address the problems of atmospheric turbulence, defocus, and field of view in a way that minimizes the need for additional hardware. One embodiment of a system for acquiring an image of a facial feature of a subject includes a single wide field of view sensor configured to acquire a plurality of images over a large depth of field containing the subject and a post processor coupled to the single sensor and configured to synthesize the image of the facial feature from the plurality of images.

Abstract: A method, apparatus, and computer readable medium for identifying a person in an image includes an image analyzer. The image analyzer determines the content of an image such as a person, location, and object shown in the image. A person in the image may be identified based on the content and event data stored in a database. Event data includes information concerning events and related people, locations, and objects determined from other images and information. Identification metadata is generated and linked to each analyzed image and comprises information determined during image analysis. Tags for images are generated based on identification metadata. The event database can be queried to identify particular people, locations, objects, and events depending on a user's request.

Abstract: Information from a position and/or gesture detection system can be embedded in a Web page, or other such presentation of content, and used to select or otherwise interact with content on the page. In some embodiments, video is captured and displayed showing a current view of the user. Position data corresponding to the video is provided and used to determine directions and extents of motion without having to do significant amounts of image processing. The position data is used to determine locations on the page where the user is attempting to provide input, such as to select an item of content. The content can be modified and/or rendered to appear to be associated with the user in the rendered view. Information from multiple gesture systems can be combined on a single page, and information from one or more gesture systems can be shared among multiple pages and devices.

Abstract: A method and system that recognize a hand gesture using selective illumination that can reliably perform hand gesture recognition by effectively removing an unnecessary image and noise including static disturbance light and dynamic disturbance light are provided. The method of recognizing a hand gesture includes acquiring, by a controller from an imaging device a hand image, which is a recognition target. A static background image and a dynamic background image are removed by the controller from the hand image. The method further includes recognizing, by the controller, a gesture of the hand by extracting a characteristic point from the hand image in which the dynamic background image is removed.

Abstract: It's an object of the invention to provide an object detection device capable of detecting an object for detection in an input image with high precision. In an object detection device 1, a detection window setting unit 11 receives a photographic image 21 photographed by a camera. The detection window setting unit 11 sets a detection window area in the photographic image 21 and generates a normal window image 22 by cutting out the detection window area from the photographic image 21. An image processing unit 12 performs image processing such as enlargement and reduction, etc. on the photographic image 21. Each modified window images 23 is cut out from each of the enlarged photographic image 21 and the reduced photographic image 21. A degree calculation unit 13 calculates matching rates indicating a possibility the object for detection in the window image for each window images on the basis of feature data 51 indicating a feature of the object for detection.

Abstract: A handwriting input system includes a handwriting pen, a handwriting plate and a processing unit. The handwriting pen includes a pen body, a first sensing unit and a first communication unit. The first sensing unit senses an action of the pen body to generate a sensing data. The handwriting plate includes a touch unit, a display unit and a second communication unit. The touch unit senses a contact trajectory and a contact time while the handwriting pen contacts the handwriting plate. The processing unit is selectively disposed in one of the handwriting pen and the handwriting plate. The processing unit calculates a tilt angle of the pen body according to the sensing data, determines a contact shape according to the tilt angle, determines an ink output amount according to the contact time, and determines a handwriting image according to the contact trajectory, the contact shape and the ink output amount.

Abstract: The present invention uses invisible junctions which are a set of local features unique to every page of the electronic document to match the captured image to a part of an electronic document. The present invention includes: an image capture device, a feature extraction and recognition system and database. When an electronic document is printed, the feature extraction and recognition system captures an image of the document page. The features in the captured image are then extracted, indexed and stored in the database. Given a query image, the features in the query image are extracted and compared against those stored in the database to identify the query image. The feature extraction and recognition system of the present invention is integrated into a multifunction peripheral. This allows the feature extraction and recognition system to be used in conjunction with other modules to provide security and annotation applications.

Abstract: A method for automated document recognition, identification, and data extraction is described herein. The method comprises receiving, by the processor, an image of a document associated with a user. The image is analyzed using optical character recognition to obtain image data, wherein the image data includes text zones. Based on the image data, the image is compared to one or more document templates. Based on the comparison, a document template having the highest degree of coincidence with the image is determined. The text zones of the image are associated with text zones of the document template to determine a type of data in each text zone. The data is structured into a standard format to obtain structured data.

Abstract: A template document and a reference document that contains objects (e.g. text objects) are received, for example, from a scanner. A plurality of common objects between the template document and the reference document are identified. Locations of the plurality of common objects in the template document and the reference document are identified. One or more distances between the locations of the plurality of common objects in the template document and the reference document are also identified. A determination is made whether the one or more distances between the plurality of common objects in the template document and the reference document are substantially the same. In response to determining that the one or more distances between the plurality of common objects in the template document and the reference document are substantially the same, the template document and the reference document are grouped as common documents.

Abstract: A forensic verification system (1100) extracts a print signature via a print signature extractor (1110) from an interior of a halftone contained in an image. The system (1100) utilizes a comparator (1120) to compare the print signature to a reference signature stored in a registry to determine differences between the print signature and the reference signature. The system (1100) utilizes a forensic analyzer (1130) to perform a forensic analysis on the signatures based on the comparison to authenticate the image.

Abstract: A computer-implemented image processing method includes accessing image data representing a plurality of images of a plurality of vehicles and reading a data model representing a predefined vehicle attribute. The method further includes processing a portion of the image data to detect a vehicle feature of at least one of the vehicles represented in at least one of the images based on the predefined vehicle attribute, and processing the portion of the image data to generate vehicle data representing the detected vehicle feature.

Abstract: A method of tracking an object includes obtaining an image captured by a camera, setting a plurality of patterns having various sizes, according to a distance from a horizon in the image to a plurality of pixels in the image, extracting an object matching one of the plurality of patterns having various sizes, while scanning the image using the plurality of patterns having various sizes, and displaying information about a position of the extracted object in the image.

Abstract: An image capturing method includes providing at least three image capturing devices arranged along a same direction and an image processor, the at least three image capturing devices capturing at least three first images, determining a target object in the at least three first images, activating a first pair of image capturing devices of the at least three image capturing devices according to shooting angles of the target object in the at least three first images in order to capture a first pair of motion images, and the image processor performing image recognition to the target object of the first pair of motion images.

Abstract: A collaborative distance metric leaning method and apparatus for visual tracking utilizes a distance metric to match the target with the best candidate. A collaborative distance metric learning algorithm is used for visual tracking by fusing two distance metrics learnt for different representations of the target. To further improve the performance of the trained distance metrics, a sample selection mechanism is also used that exploits the intrinsic structure of dense sampling.

Abstract: An apparatus includes a first acquisition unit configured to acquire main object information specifying a main object in generation of a layout image, a second acquisition unit configured to acquire object correlation information specifying an object having a correlation with the main object, an extraction unit configured to extract at least one image including the main object and at least one image including the object having the correlation with the main object from a plurality of images based on the acquired main object information and the acquired object correlation information acquired, and a generation unit configured to generate, using a layout template, a layout image in which the at least one image extracted by the extraction unit and including the main object and the at least one image extracted by the extraction unit and including the object having the correlation with the main object are laid out therein.

Abstract: A data processing apparatus may utilize an artificial neuron network configured to reduce dimensionality of input data using a sparse transformation configured using receptive field structure of network units. Output of the network may be analyzed for temporally persistency that is characterized by similarity matrix. Elements of the matrix may be incremented when present activity unit activity at a preceding frame. The similarity matrix may be partitioned based on a distance measure for a given element of the matrix and its closest neighbors. Stability of learning of temporally proximal patterns may be greatly improved as the similarity matrix is learned independently of the partitioning operation. Partitioning of the similarity matrix using the methodology of the disclosure may be performed online, e.g., contemporaneously with the encoding and/or similarity matrix construction, thereby enabling learning of new features in the input data.

Abstract: The invention allows detecting which graphics, if any, have been added to the video signal at any given time by utilizing only the various signals provided by prior art hardware. A first video signal, graphics signal(s), and a second video signal are received. It is then determined which graphics, if any, are present in a given frame of the received second video signal based on comparing actual values of pixels of the received second video signal and calculated values of pixels derived from the first video signal, and the graphics signal(s), and the combination of graphics signal(s) which produces the best match is searched for.

Abstract: A motion manifold system analyzes a set of videos, identifying image patches within those videos corresponding to regions of interest and identifying patch trajectories by tracking the movement of the regions over time in the videos. Based on the patch identification and tracking, the system produces a motion manifold data structure that captures the way in which the same semantic region can have different visual representations over time. The motion manifold can then be applied to determine the semantic similarity between different patches, or between higher-level constructs such as images or video segments, including detecting semantic similarity between patches or other constructs that are visually dissimilar.

Abstract: The invention relates to a distance-determining method by means of a camera sensor, wherein a distance between the camera sensor and a target object is determined on the basis of camera information, which method is defined by the fact that the camera information comprises a spatial extent of the region covered by the target object on a light sensor in the camera sensor.

Abstract: The invention detects a change in a subject by detecting the subject from an image, acquiring a feature quantity distribution that indicates shape information of the detected subject, accumulating the shape information that is indicated by the acquired feature quantity distribution and comparing the shape information a predetermined period of time before with the current shape information by using the accumulated shape information. Here, the invention acquires the feature quantity distribution of the subject from a processing target area extracted from an image area that includes the subject. The invention detects a change in the subject by using accumulated shape change information acquired from the shape information. The invention detects a change in the subject by using averaged shape change information obtained by averaging the shape change information.

Abstract: A method and an apparatus are used for detecting a road partition, the method comprising: a step of obtaining a disparity top view having a road area; and a step of detecting parallel lines as the road partitions from the disparity top view.

Abstract: A lane departure warning system has a trailer detector for detecting that a trailer is attached to a vehicle, a lane boundary detector for detecting a position of a lane boundary on a roadway over which the vehicle is travelling, and an information processor including a calibrator for generating calibration parameters used in determining an unintended lane departure event with respect to the lane boundary. The information processor initiates a process for allowing an operator to provide trailer width information when the trailer detector indicates that the vehicle is towing a trailer. The lane departure warning system also includes a sensor for detecting a hitch angle between the vehicle and the trailer. A controller determines a lateral offset of the trailer relative to the vehicle based on the hitch angle and a length of the trailer for generating a warning signal when the lateral offset crosses the lane boundary.

Abstract: Methods and systems are provided that may allow an autonomous vehicle to discern a school bus from image data. An example method may include receiving image data indicative of a vehicles operating in an environment. The image data may depict sizes of the vehicles. The method may also include, based on relative sizes of the vehicles, determining a vehicle that is larger in size as compared the other vehicles. The method may additionally include comparing a size of the determined vehicle to a size of a school bus and based on the size of vehicle being within a threshold size of the school bus, comparing a color of the vehicle to a color of the school bus. The method may further include based on the vehicle being substantially the same color as the school bus, determining that the vehicle is representative of the school bus.

Abstract: A detection system for vehicle operation according to an exemplary aspect of the present disclosure includes, among other things, at least one camera operable to generate an image and a display unit operable to communicate information on a display. A controller is coupled to the at least one camera. The controller is configured to command the display unit to communicate the information in response to detecting data within the image corresponding to visible light emitted in at least one predetermined frequency range. A controller and a method of detecting a traffic device are also disclosed.

Abstract: A biometric authentication device includes a photographing unit configured to photograph a biological body and acquire biological body information; an authentication unit configured to match registered biological body information that is registered in advance with the biological body information and perform authentication; a posture determination unit configured to determine steadiness of a posture of the biological body based on history of posture information relevant to the posture of the biological body when photographed; and an update determination unit configured to determine to update the registered biological body information when the posture determination unit determines that the posture of the biological body is steady.

Abstract: The present disclosure relates to a method and a system for recognizing characters. In one embodiment, the input image comprising one or more characters to be recognized is received and processed to extract one or more nodes and edges of each character in the input image. Using the extracted nodes and edges, a graphical representation and adjacency matrix of each character is generated and compared with a predetermined graphical representation and adjacency matrix to determine a match. Based on the comparison, a matching probability is determined based on which one or more characters in the input image is recognized and displayed as output. The proposed recognition method and system recognizes character with more accuracy and speed. Further, the present disclosure is simple, cost-effective and reduces the complexity involved in automatic recognition of characters.

Abstract: Techniques for determining whether touch-input gestures approximate straight lines and for animating a display with such gestures are described. Some techniques receive a gesture input through a display, define an oriented bounding box for the gesture, and determine whether the gesture approximates a straight line based on an amount of the gesture that is contained within the oriented bounding box. Other techniques determine a linear regression line for pixel locations comprising a gesture, determine distances of the pixel locations from the linear regression line, and render the set of pixel locations to the display based on the distances and a threshold. Further techniques record a gesture input through a display, determine a sequence of frame updates for the gesture, each frame update including a part of the gesture, and provide the frame updates for sequential updating of the display to effect an animation of the gesture on the display.

Abstract: The present invention provides an image processing apparatus for performing image processing of image data in which information specifying, on a first coordinate system, a position of a portion of interest in an image is recorded, including an image processing unit configured to perform rotation processing of the image data using a second coordinate system having an origin different from that of the first coordinate system, and a change unit configured to change the information specifying the position of the portion of interest in accordance with a rotation amount of the image data by the rotation processing such that an image of the portion of interest specified by the information specifying the position of the portion of interest after the rotation processing matches that before the rotation processing.

Abstract: A facility implementing systems and/or methods for creating statistically significant signatures for targets of interest and using those signatures to identify and locate targets of interest within image data, such as an array of pixels, captured still images, video data, etc., is described. Embodiments of the facility generate statistically significant signatures based at least in part on series approximations (e.g., Fourier Series, Gram-Charlier Series) of image data. The disclosed techniques allow for a high degree of confidence in identifying and tracking targets of interest within visual data and are highly tolerant of translation and rotation in identifying objects using the statistically significant signatures.

Abstract: The disclosure concerns processing of electronic images, such as hyperspectral, or multispectral images. In particular, but is not limited to, a method, software and computer for estimating shape information or a photometric invariant of a location of image of a scene. The image data (300) indexed by wavelength ? and polarization filter angle ?. For each wavelength ? index, a polarization angle ? is estimated from the image data (300) by the processor (810). The processor (810) then also estimates the shape information (such as azimuth ?, such as zenith ?, or surface normal) or photometric invariants (such as refractive index) based on the estimated polarization angle ? for each wavelength index ?. Greater accuracies can be achieved in the estimated shape information and/or photometric invariants by using wavelength-indexed data. Further, surface information or photometric invariant can be estimated based upon polarization in a single-view hyperspectral or multi-spectral imagery.

Abstract: An image processing system comprises an image processor configured to perform an edge detection operation on a first image to obtain a second image, to identify particular edges of the second image that exhibit at least a specified reliability, and to generate a third image comprising the particular edges and excluding other edges of the second image. By way of example only, in a given embodiment the first image may comprise a depth image generated by a depth imager, the second image may comprise an edge image generated by applying the edge detection operation to the depth image, and the third image may comprise a modified edge image having only the particular edges that exhibit at least the specified reliability.

Abstract: The present application is directed to new methods for automatically and effectively combining multiple metrics to select interesting, or key, frames from a video sequence. In one aspect, the methods include selecting a reference frame from a video sequence and iteratively computing the improvement or gain in the combined metrics for candidate frames with respect to that reference frame. In another aspect, a master list of reference key frames is determined, quality metrics are calculated for each of a plurality of candidate frames associated with each reference key frame, and each reference key frame in the master list is replaced with a candidate frame that yields the greatest aggregate quality improvement over that reference key frame.

Abstract: Techniques are disclosed for detecting sudden illumination changes using radiance consistency within a spatial neighborhood. A background/foreground (BG/FG) component of a behavior recognition system may be configured to generate a background image depicting a scene background. Further, the (BG/FG) component may periodically evaluate a current video frame to determine whether a sudden illumination change has occurred. A sudden illumination change occurs when scene lighting changes dramatically from one frame to the next (or over a small number of frames).

Abstract: A method for processing an image is proposed. The method comprises identifying a first group of keypoints in the image. For each keypoint of the first group, the method provides for identifying at least one corresponding keypoint local feature related to said each keypoint; for said at least one keypoint local feature, calculating a corresponding local feature relevance probability; calculating a keypoint relevance probability based on the local feature relevance probabilities of said at least one local feature. The method further comprises selecting keypoints, among the keypoints of the first group, having the highest keypoint relevance probabilities to form a second group of keypoints, and exploiting the keypoints of the second group for analyzing the image. The local feature relevance probability calculated for a local feature of a keypoint is obtained by comparing the value assumed by said local feature with a corresponding reference statistical distribution of values of said local feature.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network to detect object in images. One of the methods includes receiving a training image and object location data for the training image; providing the training image to a neural network and obtaining bounding box data for the training image from the neural network, wherein the bounding box data comprises data defining a plurality of candidate bounding boxes in the training image and a respective confidence score for each candidate bounding box in the training image; determining an optimal set of assignments using the object location data for the training image and the bounding box data for the training image, wherein the optimal set of assignments assigns a respective candidate bounding box to each of the object locations; and training the neural network on the training image using the optimal set of assignments.

Abstract: Embodiments of the invention provide a method for scene understanding based on a sequence of image frames. The method comprises converting each pixel of each image frame to neural spikes, and extracting features from the sequence of image frames by processing neural spikes corresponding to pixels of the sequence of image frames. The method further comprises encoding the extracted features as neural spikes, and classifying the extracted features.

Abstract: Systems and methods for test object classification are provided in which the test object is docked with a target object in a plurality of different poses to form voxel maps. The maps are vectorized and fed into a convolutional neural network comprising an input layer, a plurality of individually weighted convolutional layers, and an output scorer. The convolutional layers include initial and final layers. Responsive to vectorized input, the input layer feeds values into the initial convolutional layer. Each respective convolutional layer, other than the final convolutional layer, feeds intermediate values as a function of the weights and input values of the respective layer into another of the convolutional layers. The final convolutional layer feeds values into one or more fully connected layers as a function of the final layer weights and input values. The one or more full connected layers feed values into the scorer which scores each input vector to thereby classify the test object.

Abstract: An electronic card connector (100) includes an insulative housing (1), a number of terminals (2) retained in the insulative housing, a movable component movable between an original position and a locking position, a shielding shell (4) attached to the insulative housing and having a top wall (41) and a first elastic portion (43) located at the top wall, and an ejecting mechanism (5). The ejecting mechanism has a lever (51) having a resisting part (5111) and a sliding bulge (54) movable together with the movable component. The first elastic portion has a locking part (431) resisting against the sliding bulge along an up-to-down direction at the locking position and a guiding part (432). The lever is movable forwardly and the resisting part is guided by the guiding part to lift the first elastic portion to thereby disengage the locking part from the sliding bulge.

Abstract: A non-transitory computer-readable storage medium storing a printer driver includes a printer driver core component and a print setup user interface component. The print setup user interface component causes the computing device to control a display section to display one or more setup user interface portions by determining whether a setup user interface portion designated by designation information obtained from the printer driver core component is implemented in the print setup user interface component; in response to determining that the designated setup user interface portion is not implemented, searching for a compatible setup user interface portion; and in response to determining that the designated or compatible setup user interface portion is implemented, controlling the display section to display the designated or compatible setup user interface portion.